How can retailers maximise store performance?

Print

In late 2017, Nordstrom, an American chain of luxury department stores, started a retail concept called ‘Local’ to better engage its customers with new offerings. It offers a host of services like fashion consulting from a personal stylist and manicures, with the revolutionary idea that the stores carry no inventory. The company’s innovative approaches are being credited for its better-than-expected financial performance.

By way of stark contrast, Toys R Us filed for bankruptcy last year and has now shut down operations due to a combination of factors – failing to keep up with consumer demands, and mounting margin pressures.

 

Retailers today face a two-pronged challenge:

  • on the one hand, they are striving to stay ahead of the competition by offering superior customer experience, and products and services that address new demands and trends;
  • on the other hand, they have to maintain financial discipline, reduce costs and be profitable if they are to survive.

 

“Where to start?”

Nordstrom’s ‘Local’ concept faces the same challenges as other retail innovations. Companies have to be selective in which services they offer, and cannot be overly ambitious or take too much risk, such as rolling out new initiatives across all locations in one go.

Retail clients have been increasingly asking us how to identify which products, services, or offerings to introduce and where. Take the example of a large European grocery chain we work with on an ongoing basis. It had experimented opening sushi bars in a few of its stores to both delight customers and increase revenue. As the initial trial yielded success, the company wanted to replicate it more widely. The major caveat in such interventions is that a retail chain with 1000+ stores cannot take the risk of investing in a sushi bar everywhere and simply hope for it to succeed. It is too big a financial risk. To allocate capital appropriately, the retailer wanted to identify the drivers of sushi bar performance so that it could extend the experiment to stores that had the best opportunities for success. The question of “where to start” becomes really important in these sorts of new retail offerings.

 

The true ROI

Data analytics plays a key role in predicting return on investment for specific products and services in particular stores. Here are three strategies to maximise retail ROI:

 

1. Identify and understand the business problem

Retailers collect huge amounts of data. Walmart, for example – the world’s biggest retailer with over 20,000 stores in 28 countries – announced in 2017, that it was in the process of building the world’s biggest private cloud, to process 2.5 petabytes of data every hour.

It is not uncommon to lose focus when there is an abundance of data. After all, data is being gathered from every imaginable source: a consumer’s past purchases, store visit frequency, online search history – even visuals from store cameras. All this information can be useful in enhancing customer experience and maximising store performance, if the retailer asks the right questions. For this, the key is to articulate a structured hypothesis and then test it. The better the articulation of the problem, the more effective the solution is likely to be.

 

2. Use data analytics to find customised solutions for each store

What we have found and measured by using predictive analytics for retailers, is that product and service offerings have varying success rates in different stores. A premium salad dressing may sell well in a store in New York, but may not have the same appeal to customers in a Minnesota store. A customised approach is recommended when selecting stores for new offerings, using data-based predictions.

One of our large global retail clients wanted to investigate the likely success of stocking a premium bakery brand. As the products also have premium prices, a surplus would put cost pressures on stores. However, with the right number of SKUs, and in areas where there is demand for such offerings, a store’s sales could be improved. The retailer wanted to know which stores had the highest likelihood of generating sales. The Smart Cube used statistical modelling techniques (such as regression) to identify the drivers of product sales in test stores.

In this case, one of the top influencing factors turned out to be the proportion of customers aged 45+ living within a 10-minute drive of the stores. Would the retailer have been able to make this observation without supporting data? Probably not; especially when the underlying assumption was that customers in their 20’s and 30’s would be more likely to buy premium bakery products. Data analysis challenged this assumption, providing the business with the insights needed to develop a successful product strategy. The result? The 150+ stores which rolled out the new bakery brand achieved the sales target for the new bakery lines.

 

3. Embrace experimentation to drive innovation

Innovation needs to be of upmost importance for retailers, if it isn’t already. Data-driven experimentation can accelerate the use of data and advanced analytics to improve store performance. Experiment design, as a methodology, requires retailers to first establish the business problem/opportunity, then pilot and test on a selection of trial stores. The business problem can be investigated in a structured and therefore limited time- and cost-basis. Using an agile methodology and sprint-based delivery principles, retailers can experiment, make mistakes, learn from them, and build the learnings into the process.

Our retail analytics team recently used the experiment design methodology to help a large European retailer. The client’s business problem was how to make customer checkout a faster and more simple experience. To achieve that, the retailer wanted to optimise the number and configuration of its checkout counters. Starting with the principle that a particular store had an optimum 13 counters, the client wanted to know how many would need to be conventionally manned, self-checkout with cash and/or card payment options, and ‘smart shop’ checkout. Our team applied the experiment on a select few stores, using our Concept Lab methodology. This entailed conducting sprints to do analytical data preparation, exploratory analysis and logistics regression modelling. Once the drivers of customers’ checkout counter preferences were identified based on the experiment’s results, the client rolled out the recommendations in 20 stores.

 

What does the future hold?

Retailers face stiff competition and constantly evolving consumer expectations. Many have and will go bust if they don’t find ways to adapt – in late 2018 Sears joined 14 other US-based retailers which filed for bankruptcy over the course of the year. Yet some are innovating and succeeding – IKEA for example is using augmented reality to give a differentiated experience to its customers, and continues to open new stores world-wide. Which bracket do you want to be counted in?

 

Find out more about The Smart Cube’s solutions for retailers today, and how analytics can help ease increasing margin pressures.

 

With additional inputs from Angad Bhatia, Arun Kumar and Sameep Rohatgi.

 

Raman Sharma

Raman Sharma

Raman spearheads the retail analytics practice for The Smart Cube. In his role as Associate Vice President, he develops analytical solutions for business teams within of retail clients, including marketing, CRM, and operations. Outside of work, Raman enjoys reading books, watching cricket and spending time with his family.

See more

Ready to talk to us about the challenges facing your business?